一种估计和检验排序模型中结构变化的方法及应用

发布时间：2018-06-24 06:28

本文选题：平滑转移 + 排序变量　；参考：《数量经济技术经济研究》2017年08期

【摘要】：研究目标:给出一种估计和检验排序模型中结构变化的方法。研究方法:在排序模型中引入平滑转换函数来描述结构变化,在此基础上构建拉格朗日乘子统计量检验模型中的结构变化,并使用极大似然方法估计模型。研究发现:拉格朗日乘子统计量具有标准的渐近卡方分布,并且该统计量对误差项的不同分布形式具有较好的稳健性;极大似然估计量具有一致性和渐近正态性;应用本文的方法分析居民收入和幸福关系,发现收入对幸福的影响存在显著的非线性结构变化特征,当收入增长超过社会收入分布的80%分位数时,收入对幸福的作用会减弱。研究创新:提出了一种新的并且是比较简便的估计和检验排序模型中结构变化的方法。研究价值:这种新的方法可以广泛应用于主观评价问题中的结构变化分析。
[Abstract]:Objective: to provide a method to estimate and test the structural changes in the sorting model. Research methods: a smooth transformation function is introduced to describe the structural changes in the ranking model. On this basis, the Lagrange multiplier statistics test model is constructed, and the maximum likelihood method is used to estimate the model. It is found that the Lagrange multiplier statistic has the standard asymptotic chi-square distribution, and the statistic has good robustness to the different distribution forms of the error term, the maximum likelihood estimator is consistent and asymptotically normal. By analyzing the relationship between income and happiness in this paper, it is found that the influence of income on happiness has significant nonlinear structural characteristics. When income growth exceeds the 80 quartile of social income distribution, the effect of income on happiness will be weakened. Research and innovation: a new and relatively simple method for estimating and examining structural changes in sorting models is proposed. Research value: this new method can be widely used in structural change analysis of subjective evaluation problems.
【作者单位】：中南财经政法大学统计与数学学院;
【基金】：教育部人文社科基金;教育部留学回国人员科研启动基金项目资助
【分类号】：F224.0

【参考文献】